Oracle Database Discussions

Hi All,

We have got some huge partitioned tables. Currently our client is using partname =>'partition_name' instead of granularity option in dbms_stats.gather_tables stats. Now client is complaining that job takes hugh time to complete(gather_table_stats), earlier we used gather_schema_stats but that didnt help either.

Current job is -

1)dbms_stats.gather_tables_stats(owname =>'owner',tabname=>'table name',partname =>' partition name', degree => 28, granularity =>'PARTITION').

Here client is not using estimate_percent and cascade.

I want to use -

2)dbms_stats.gather_tables_stats(owname =>'owner',tabname=>'table name', degree => 28, granularity =>'ALL', increment =>'TRUE', estimate_percent =>10);

So will 2nd statement will run in less time as compared to 1st? Remember its 11.1.0.7 release. And we have option like 'increment statistics'.

Suggestion will be highly appreciated.

Regards
Ranjit